AP PHOTO/MARK LENNIHAN PHOTO/MARK AP Unbundling ‘Too Big to Fail’ Why Big Is Bad and What to Do About It By Ganesh Sitaraman July 2014 WWW.AMERICANPROGRESS.ORG Unbundling ‘Too Big to Fail’ Why Big Is Bad and What to Do About It By Ganesh Sitaraman July 2014 Contents 1 Introduction and summary 2 Unbundling too big to fail 9 Reforming too big to fail 18 Conclusion 20 Endnotes Introduction and summary Since the 2008 financial crisis, the problem of financial institutions being “too big to fail,” or TBTF, has been front and center in the public debate over the reform and regulation of the financial industry. Commentators across the political spectrum decried bailouts of the biggest Wall Street financial institutions, arguing that bail- outs would establish too big to fail as public policy. When it was time for reform, legislators tried to address this problem, and even incorporated into the full title of the Dodd-Frank Act that one of the bill’s purposes was “to end ‘too big to fail.’”1 Yet more than five years after the financial crash, the biggest banks are 37 percent larger than they were before,2 and the debate over what to do about the size of finan- cial institutions continues. Policy proposals range from improving resolution mecha- nisms, to more stringent prudential standards such as leverage limits, to charging fees to eliminate the implicit government subsidy the biggest banks receive, to capping the size of the banks, to instituting a new Glass-Steagall Act. Each approach is hotly contested, with commentators frequently arguing that the proposed solution will not actually fix the problem of financial institutions that are too big to fail.3 The problem at the heart of the debate over too big to fail is that the popular moniker has come to mean more than the concern that big firms get a government bailout in the event of failure. It captures a variety of concerns with the financial industry: eco- nomic, competitive, systemic, firm level, political, legal, and regulatory. This report identifies the full range of reasons reformers might be worried about TBTF. It then describes the various policy options that are most frequently discussed with regard to reforming TBTF, and it connects the specific reforms to the concerns they address. To make progress, reformers and critics alike need to engage in a more precise debate. Critics too often dismiss reforms without fully addressing the concerns reformers seek to address, and when outlining proposals, reformers could be clearer about the problems they seek to solve. Ultimately, people will disagree about what aspects of TBTF are most concerning to them. But the first step toward a more meaningful debate over reform requires greater clarity about the particular concern—or concerns—with big financial institutions and the specific solutions that address those concerns. 1 Center for American Progress | Unbundling ‘Too Big to Fail’ Unbundling too big to fail The underlying concerns with big financial institutions can be grouped into five categories: systemic risk, bailout, competition, firm-level concerns, and political control. Each incorporates multiple related worries about the pernicious effects of the size of financial institutions. Systemic-risk concerns While there are a variety of definitions of systemic risk, the basic concept is simple: systemic risk exists when the failure of a single institution would have significant effects beyond the firm, to the financial system or the economy as a whole.4 These ripple effects are seen as so troubling that action must be taken to prevent them, whether that means bailouts after firm failure or some form of regulation before failure. Size-based systemic risk The concern about size-based systemic risk—the most natural reading of too big to fail in the systemic-risk context—is that the failure of a gigantic financial institution will have immense effects on the financial system or the economy as a whole, simply because the firm is extremely large. Size in this context is obviously a proxy for importance, albeit an imperfect one.5 Large institutions might be able to fail without harmful systemwide effects. Likewise, smaller institutions that are central to the functioning of the system might fail with disastrous consequences. But objections on these grounds are largely a debater’s point. The argument is that size is a reasonable and relatively workable proxy for importance. 2 Center for American Progress | Unbundling ‘Too Big to Fail’ Interconnectedness-based systemic risk Many argue that TBTF should instead be called “too interconnected to fail,” a description that identifies a different kind of systemic risk.6 The worry here is that an institution has too many links to other institutions in the economy, such that its failure would have negative effects throughout each of these firms and thus to the system as a whole. This form of risk is often called “contagion,” as the “disease” in one institution will spread to others with which it interacts.7 For example, if firm A is engaged in risky behavior and fails, it may not be able to fulfill its contracts with firms B and C, each of which then fail, affecting firms D and E, with whom they work, and so on. This cascade effect ripples through the economy. Of course, interconnectedness is not limited to financial institutions. The bail- out of General Motors, or GM, in 2008 and 2009 was in part justified by links between GM and its parts manufacturers, distributors, and others in the automo- tive industry.8 Similarly, imagine the systemic effects of a hypothetical Wal-Mart failure: The firm’s connections throughout the consumer-goods industry would have significant effects on major consumer-product companies. System effects and systemic risk A variety of systemic-risk issues also arise when multiple actors operate within a single system whether or not they are interconnected. Three main concerns arise. The first is informational.9 If firm A fails, people may scrutinize firm A’s risky prac- tices, only to realize that firm B has been engaged in the same practices. If people believe those risky practices led to failure in firm A, they may no longer want to work with firm B because it engages in the same risky practices. The second con- cern is often called “common shock,” defined as a situation in which a single exter- nal event affects multiple firms at once, leading to the failure of the institutions simultaneously.10 The third concern is the conventional notion of a panic—that the failure of a TBTF firm will lead to widespread public panic that undermines multiple firms, regardless of the soundness of those other firms.11 One phenomenon that might undergird each of these systemic risks is the “too- many-to-fail” concern, or so-called herd mentality within the industry.12 Economists have shown that when there are isolated bank failures, regulators will allow other institutions to acquire the failing banks, but when there are many simultaneous 3 Center for American Progress | Unbundling ‘Too Big to Fail’ bank failures, regulators find it optimal to bail out some or all of the failing banks. As a result, smaller banks “herd” toward the policies of large banks to benefit from the bailout policy in the event of failure. This alignment means that multiple bank failures—and, as a result, bailouts—are more likely to occur at once. Bailout concerns Another common area of concern is taxpayer bailouts of failing firms. The worry is that the government will bail out firms that are seen as TBTF instead of letting them fail as the principle of creative destruction requires. Whether explicit or implicit, a policy of bailouts skews incentives for firms and harms taxpayers and markets as a result. Bailout recipients can differ—management, shareholders, or creditors—but regardless, bailout concerns focus on two specific issues: moral hazard and cost. Moral hazard First is the idea that bailouts lead to a moral hazard.13 TBTF firms know that they can benefit from government bailouts in the event of staggering losses. As a result, it is rational for them to take on riskier behavior because they will be able to capture the profits while socializing the losses among taxpayers. The result is a system that fosters more and more risky behavior because the government’s bailout policy has undermined the disciplining effects of the downside risks that accompany market participation, such as losses, failure, or acquisition. Bailout costs Bailouts also mean that taxpayers are on the hook for losses from the risky bets of private actors. If bailouts become a recurring practice, it is not obvious that this will result in financial returns for the U.S. Treasury. But even if the costs of bailouts are paid back to the Treasury over time, the practice is troubling for both moral and practical reasons. Morally, taxpayers should not have to rescue those financial institutions taking on risky activities that have questionable social value—includ- ing giving bonuses or golden parachutes to top executives whose actions caused their institution’s failure. Practically, in a world of constrained resources, there might be opportunity costs to spending money on bailouts for the largest financial 4 Center for American Progress | Unbundling ‘Too Big to Fail’ institutions during a financial crisis, rather than on economic stimulus policies or pro-growth policies such as investment in infrastructure or research and develop- ment. If Congress feels budgetary constraints, then prioritizing bailouts might mean that other important spending gets short shrift. Competition concerns TBTF also includes a variety of concerns about competition within the financial industry in which the biggest firms gain undue advantage over smaller firms.